Neural Network Applied On The Detection Of Harmonics And Voltage Sag

Power is the most widely used energy in the modern society. Its application extent is one of the important indicators of the development level of a country. In recent years, under the rapid development of China's power industry, the size of the power system is increasing. Power shortage problem has been gradually resolved, but at the same time, the users want to improve the power quality; power quality problems get more and more attention. The power system harmonics is one of important factors on affecting power system security and economic operation; and voltage sag is the most important industrial power quality problems. How to ensure the power quality effectively and ensure the normal operation of power systems are the problems to be solved quickly. In order to obtain high-quality energy, the detection of power quality characteristic is particularly important, so kinds of detection methods have been developed rapidly.1) This article first described a variety of related issues about power quality, the harmonic, voltage sag and so on in power system. Analyzed kinds of methods of detection on the harmonic and voltage sag, specially. Introduce systematically the advantages and disadvantages of these methods.2) The paper described the artificial neural network on harmonic detection and voltage sag detection, and described the application on the detection of power quality characteristics of the neural network method. Presented a feedback neural network model based on sine basis function, and tested the stability of this model.3) Finally, the model was simulated by MATLAB simulation software, and simulated the harmonic and voltage sag, detected a variety of output value. Analyzed the simulation results, the model had a good results in real time and accuracy by simulation graphics and simulation data.